Building energy performance simulation: a case study of modelling an existing residential building in Saudi Arabia

Internationally, a broad range of scientifically validated building performance simulation (BPS) tools are accessible. Computer simulation software tools have made major strides forward in recent years and will continue that trend since they make the evaluation of the whole process of design, operation, maintenance, and lifecycle processes of any building possible from concept to design [18]. Furthermore, these tools can be utilised to assess the energy performance of an existing or new building. Furthermore, BPS can integrate human activity and thermal and visual comfort simulations into the computer modelling and simulation process of buildings [19, 20]. Previous literature [18, 21, 22] have mentioned some of the noticeable advantages of the application of computer based simulation and modelling. A study done by [23] that outlines major criteria for BPS tools evaluation and selection based on analysing user's needs for tools capabilities and requirement specifications concludes that DesignBuilder (DB) is almost the only tool that was appreciated by architects and engineers and was ranked in the top. DB is a building simulation software that was created specifically for running EnergyPlus simulations on virtual building models [24]. Other studies [24–26] have emphasised the validity and suitability of DB software, especially in the context of building performance studies. As for this reason, this study will employ DB as the main BPS tool.

The case study analysis method can be used to elucidate minute facts and in-depth details about real phenomena [27]. The most dominant dwelling types according to the public survey were flats and detached two-storey villas. However, flat buildings are mainly constructed for investment purposes and therefore they may be difficult for in-depth study. Hence, the building type that has been selected to be analysed in this study is the detached house type. The building has to be existing and occupied to allow for data gathering including building specification, users profiling, and energy used. This study will focus on the prevailing climatic zone in Saudi Arabia (zone 1) as according to the SBC classification. It contains regions characterised by the hot dry climate which comprise almost 65% of the area in the KSA. The city of Riyadh was selected to be the representative study area.

Riyadh, capital city of the KSA, is located in Riyadh region at latitude 24.7°N and longitude 46.8°E and at about 600 m above the sea level sloping eastward. Since the past five decades, the city of Riyadh began to change from a small-enclosed town to a modern city covering an area of 3115 square kilometres, including 15 municipalities and home to more than 6.5 million people in 2017 [28]. Because of its economic growth and availability of job opportunities, Riyadh has become a magnet for people from other regions. As a result, population growth has risen, as has the need for more housing. According to the Saudi housing statistics 2019, the number of houses occupied by Saudi citizens in Riyadh was 865.4 thousand, a share of 23.5% of the total houses number in the KSA [29]. Also, the average family size in the KSA is 5.86. Most dominant housing unit in Riyadh is villa type contributing to about 46% of the total houses in Riyadh.

It is usual for Saudi Arabian summer ambient temperatures to go higher than 46.1 °C with mean monthly temperatures range between 27.3 °C and 37.1 °C, see figure 3. The extremely high summer ambient temperatures require suitable cooling systems to maintain thermal comfort for building occupants [30].

Zoom In Zoom Out Reset image size

The model parameters must be specified before the DesignBuilder software can simulate a house's thermal performance. Physical characteristics such as geometry and plan, installed appliances or equipment, building function and occupancy profile, site location and climate, and the nature of the surrounding environment, among other things, are defined by model parameters [18]. This definition is required in order for the DB tool to select the appropriate material for modelling. Models in DB are arranged in a basic hierarchy. This arrangement allows to create settings at the building level that become active in the entire building [31]. Default data is inherited from the level above in the hierarchy.

The case study building used for this thesis is a two-storey detached villa, a conventional building representing the typical Saudi dwellings. Table 1 represents the main characteristics of the selected house.

The ground floor of this building consists of separate guest room for each gender, guest dining room, common lounge, kitchen, and two toilets. This design style is very popular in the KSA which reflects Saudi Muslim culture, that is mainly involving gender separation. Furthermore, main bedrooms are located in the upper floor, figure 4. The household of the building used in this study indicated that there are some rooms on the ground floor that are occasionally occupied such as guest rooms and guest dining room. The upper floor, on the other hand, seems to consume most of the end-use energy due its continuous occupancy.

Zoom In Zoom Out Reset image size

As stated earlier, the residential building sector consumes a substantial amount of energy relative to other major energy consumers, and this is expected to increase in the future if the public remains unaware of the country's current energy situation and sustainable building energy alternatives. Energy bills are important to this study because they specify the rate of energy usage in buildings and the amount of money spent on energy by householders on a monthly or annual basis. Hence, the monthly electricity bills for year 2019 were gathered from the household and the Saudi Electricity Company. The total annual energy consumption of the selected building was 46 143 kW h (114.5 kW h m⁻²).

Typically, houses in Saudi Arabia are masonry structure made from reinforced concrete and concrete blocks. The construction of the envelope of the examined building is without insulation, according to the household. The construction details of the building's envelope are given in appendix A. This practice seemingly still common in residential buildings construction in Saudi Arabia. The modelling drawings and simulation inputs are shown in figure 5 and table 2, respectively.

Zoom In Zoom Out Reset image size

Having the parameters defined using input data given in table 2 and the building elements construction details, the simulation output reflects the total annual energy consumption in kW h. A breakdown of the monthly energy use in kW h (i.e., electricity by rooms, lighting, heating, cooling, and domestic hot water) is displayed in figure 6. The demand for water heaters across the region to satisfy rising domestic applications of warm water such as cooking, cleaning, and bathing has also contributed to the increasing energy consumption in residential buildings. It is worth noting that the computer simulation was able to capture the considerable rise in energy consumption from May to October, owing to the use of air conditioning during this period. The annual energy consumption of the modelled base case is 47 389 kW h (117.6 kW h m⁻²). It is obvious from figure 7 that most of the total annual energy consumed (82%) was attributed to cooling loads, 38 836 kW h (96.4 kW h m⁻²). This is justified by the local harsh climatic condition. The total energy consumption for space cooling in kW h m⁻² per year is 96.2. The European Standard recommended 20–30 kW h m⁻² per year space cooling [32].

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

In addition, the total CO₂ emissions in kg for the whole year is 28 718. The emissions increased in summer months due to the operation of cooling systems to maintain occupants' thermal comfort at desired level from an environmental perspective. As there are six occupants in this house which agrees with the size of the average family in [33, 34], the CO₂ emission per capita is about 4786 kg (4.8 tonnes), which is almost twice the average CO₂ emission per capita of the 25 EU member states (about 2.5 tonnes) [35]. High energy consumption results in high CO₂ emissions, and vice versa. Environmental protection can be obtained if more sustainable domestic buildings are designed.

The thermal comfort of occupants is difficult to analyse because it is as much psychological as it is physiological. The term 'comfort' is a state of mind or a personal feeling—not a quantifiable metric [36]. However, Fanger has established a model to evaluate occupants' comfort with account to specific parameters. He extended the usefulness of his work by proposing a method by which the actual thermal sensation could be predicted by producing values for the predicted mean vote (PMV) and the predicted percentage of dissatisfied (PPD) [36].

Acceptable PMV and PPD ranges:

• ASHRAE 55 standard states that the recommended thermal limit on the seven-point scale of PMV is between −0.5 and 0.5 with a corresponding PPD falling below 10%.
• ISO 7730 expands on this limit, giving different indoor environments ranges. ISO defines the acceptable comfort limits range between −0.7 and +0.7 for old buildings, and between −0.5 and +0.5 for new buildings.

EnergyPlus provides a sophisticated building thermal analysis tool that allows to determine if the environmental control strategy would be sufficient to maintain occupants thermal comfort [31]. This tool is the DesignBuilder thermal comfort calculator. The outcome of thermal comfort calculator is reflected by Fanger comfort model shown in figure 8.

Zoom In Zoom Out Reset image size

Figure 8 indicates that the comfort analysis in the existing building falls outside the comfort limits with a PMV value of −1.48 and a PDD of 49.9% which implies the fact the indoor condition is cool. This can be justified with the lack of thermal insulation and hence the continuous operation of air conditioning to level the discomfort caused due to hot outer weather of Riyadh city. This is in line with [36] where it states that hot climatic region in summer season have a trend of −1 and −2 actual thermal sensation vote which shows that occupants are overcooled.

Validation ensures that research is conducted in an objective and unbiased way [37]. Accordingly, and to prove that the study is close to reality and representative of the actual building, a comparison of values between gathered real life energy consumption and DB simulated cases was made as shown in figure 9.

Zoom In Zoom Out Reset image size

Due to increased levels of automation, lower costs, and other factors, the whole-building approach of building energy models (BEMs) is now more popular than the single-measure approach [38–41]. Since the BEM accuracy is a deciding factor in all applications, calibrated models are needed. The task of determining the accuracy of BEMs is critical since once the model has been validated through a calibration process, it can be used to test and implement various strategies for reducing energy use while preserving human comfort. Calibration is defined according to ASHRAE 14-2014 guidelines as the: 'process of reducing the uncertainty of a model by comparing the predicted output of the model under a specific set of conditions to the actual measured data for the same set of conditions...' [42]. The three principal guidelines that clarify how to determine this 'degree' of confidence, its uncertainty, are FEMP [43–46], ASHRAE guideline 14, and IPVMP [47–50], summarised in table 3.

The principal uncertainty indices used are: normalized mean bias error (NMBE) found by equation (1), coefficient of variation of the root mean square error (CV(RMSE)) which measures the variability of the errors between measured and simulated values as in equation (2), and coefficient of determination (R²).

$$\begin{equation}\mathrm{N}\mathrm{M}\mathrm{B}\mathrm{E}=\frac{1}{\bar{m}}\cdot \frac{\sum\limits _{i=1}^{n}\left({m}_{i}-{s}_{i}\right)}{n-p}\times 100\left(\%\right)\to \bar{m}=\frac{\sum\limits _{i=1}^{n}\left({m}_{i}\right)}{n}.\end{equation} \tag{ 1 }$$

Where: m_i refers to measured value, s_i refers to simulated value, n is the number of measured data points (in this study n = 12 months), (m˜) is the mean of measured values, p is the number of adjustable model parameters, which is suggested to be zero, for calibration purposes.

It is worth noticing that positive values mean that the model under-predicts measured data, and a negative one means over-prediction. ASHRAE guidelines [42] subtract measured values (m_i ) from simulated ones (s_i ) instead of FEMP [44, 45] and IPMVP [49], which do the opposite. For this reason, the explanation of the under- or over-prediction is inverted. In equation (2), the value of p is suggested to be one [50, 51]. Table 4 shows the final calculations for validating the model using both equations.

$$\begin{equation}\mathrm{C}\mathrm{V}(\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E})=\frac{1}{\bar{m}}\sqrt{\frac{\sum\limits _{i=1}^{n}{\left({m}_{i}-{s}_{i}\right)}^{2}}{n-p}}\times 100\left(\%\right).\end{equation} \tag{ 2 }$$

Coefficient of determination (R²) indicates the proximity of simulated values to the regression line of the measured values. It is a statistical index that is widely used to calculate the uncertainty of a model. It is limited to a range of 0.00 to 1.00, with the upper limit indicating perfect match of simulated and measured values while the lower limit indicating the opposite. Both the ASHRAE Handbook [52] and the IPVMP [49] suggest that it should never fall below 0.75. In this study, R² value was equal to 0.987 as depicted in figure 10.

Zoom In Zoom Out Reset image size

It is obvious, according to the above validation criteria in table 3, that this model has met these guidelines and proven to be valid and indicates that the predictions from monthly simulation highly fit the data from the actual measurements.

Since most Saudi buildings heavily rely on mechanical cooling systems to ensure thermal comfort, a proper strategy for reducing the energy consumption is required. This included: (1) improvement of thermal resistance in building envelopes (walls and roof); (2) application of advanced window systems (window to wall ratio WWR, glazing type, and shading devices); (3) airtightness expressed by infiltration rate (i.e. air changes per hour ac/h); (4) the use of energy efficient mechanical systems (cooling setpoint temperature, and AC coefficient of performance CoP). The DB parametric analysis tool enables to see the effect of different parameters on the building's annual energy performance (in kW h). This process allows for comparing different inputs for each individual parameter in term of total annual energy consumption. For the purpose of this study, simulation parameters (also called variables) were categorised into: design-related and behaviour-related parameters described in table 5. These parameters, individually and as a whole set, were analysed in order to find their impact on the total energy consumption.

A detailed parametric analysis of the prototypical model can be used to determine the effect of many design and operating measures on the total annual energy consumption. Figure 11 depicts the total energy saving percentage through the complete simulation associated with all individual parameters (stated in table 5) where it can be seen that the improved case impact increases the energy savings.

Zoom In Zoom Out Reset image size

As expected, figure 11 indicates that adding insulation has the most significant impact, reducing annual consumption by 22.8% and 18.4%, for walls and roof, respectively. It is worth noticing that the lower the U value of the insulation composition the better thermally the insulation will be leading to greater energy saving. This is mainly by the effect that thermal insulation materials pose on the heat gain through envelope elements which subsequently affects the dominant cooling load in a positive way. The combined addition of thermal insulation in building's walls and roof can achieve 31% to 45% savings in total energy consumption. This is consistent with other studies, which suggested energy reductions of 15% to 35% when complying with exterior walls and roofs thermal insulation requirements [53–56].

The measure with the second highest impact is installing an energy efficient air conditioning (i.e., CoP) which can result in 15% reduction of the annual energy consumption. According to SEEC, most common AC types used in dwellings are windows systems, and/or windows combined with split systems. Old versions of window AC (CoP = 2.0 to 2.5) are considered inefficient with respect to energy consumption [57].

Increasing the WWR negatively affects the energy consumption and leads to no saving as shown in figure 11. This falls in line with [14] where adopting a WWR of 40% led to 32% increase in the total energy use. However, a study done by [33], reported that only a 2% difference in the annual energy use was observed between the WWR of 25% and 50% which reflects the minimum overall effect on energy consumption.

The greater the airtightness at a given pressure difference across the envelope, the lower the infiltration [31]. The Federation of European Heating, Ventilation and Air Conditioning Associations (REHVA) suggests that there are a growing number of studies indicating that there is considerable impact on energy use in buildings in mild and hot climates [12]. For the purpose of this study, the reference building is assumed to have an infiltration rate of 1.0 ac/h. Reducing this rate to 0.25 ac/h will lead to 12% saving. Therefore, it is highly recommended to design and construct the building fabric to be reasonably airtight and to seal air leakage sources around building envelope as this positively contribute at maintaining good indoor air quality and minimising energy use [12].

Finally, minimum impact on energy saving was noticed when considering other energy efficiency measure (EEM) such as increasing the cooling set temperature, installing double glazing windows, and utilising local shading. Interestingly, the effect of two behaviour related parameters, cooling setpoint temperature and CoP, contributed to an annual energy saving of 18%. This is to quantify the overall effect of occupants' behaviour related variables on total energy consumption. The Saudi Energy Efficiency Centre (SEEC) recommends the set thermostat point of air conditioning be between 23 °C and 25 °C [57] although in most cases occupants still tend to set their internal cooling temperature at 18 °C [58].

Many countries around the world are using regulation for new buildings to produce minimum values of thermal insulation, with an analysis of the economic and environmental impact on the country [59–62]. As previously mentioned, the 2030 vision aims to control the high load demand in several ways, like making all buildings efficient. Hence, the SBC-602 is being eventually reinforced by the Saudi government. The main goal of SBC-602 is to decrease the cooling and heating load with optimum thermal insulation. In the past, the SEC has initiated a set of minimum requirements of thermal insulation to achieve efficiency in new and existing buildings as well. Table 6 represents the requirements for thermal insulation in local authority codes (SBC-602 and SEC) which only consider the thermal characteristics of building envelope members (wall, roofs, and glazing's (W m⁻² K). It also stated those characteristics for both base and improved case.

This section will only evaluate the impact of applying thermal insulation requirements in zone 1 considering the vast area covered by climate zone 1 which is the most significant hot zone in Saudi Arabia [63]. To draw conclusion, the residential building will be studied in the following four scenarios:

• Not insulated (existing and relatively old buildings i.e., base case).
• It complies with the SEC construction specifications.
• It fulfils the SBC-602 thermal insulation requirements.
• It applies those parameters with the highest effect on energy demand (as in figure 11)

This section will only focus on the impact of improving the envelope insulation in reducing energy consumption in the residential sector. Hence, only the parameters mentioned in table 6 will be considered in the following analysis. Note that the basic case for comparison is at a setpoint temperature of 20 °C. The outputs of the studied four scenarios are summarised in table 7. By comparison between the base case energy used intensity (EUI) value and the results presented in the work of [64] for Riyadh, where the value ranged from 100 to 162 kW h/(m² year), it was found that the result obtained falls within the range that was obtained previously [33, 64, 65].

By comparison between these values and the results presented in table 7, it is found that the annual electrical energy when applying the minimum requirement of thermal insulation by SEC is still considered high when compared with SBC-602 and best improved scenarios. An annual energy reduction of 46.2% and 48.1% was observed when applying SBC-602 and improved case thermal insulation, respectively. SBC-602 and improved scenarios seem to be identical in terms of building energy performance where the annual saving of electricity consumption will be about 21 888 kW h and 22 787 kW h, respectively. The comparison of the total monthly electrical energy consumption at 20 °C can be seen in figure 12.

Zoom In Zoom Out Reset image size

Figure 13 benchmarks the annual energy consumption per unit area of the existing case with different standards and results obtained by other researchers. The energy consumption value investigated in this study for the improved case is lower than the benchmark value for energy-efficient residential buildings in Norway, France, and Germany, which is, for instance, less than 70 kW h m⁻² [66]. These standards reflect the importance and significance of efficiency requirement in domestic houses. It is worth mentioning that the analysis in section 3.5 only considered the application of proper insulation only and not the whole performance parameters which is discussed in section 3.6.

Zoom In Zoom Out Reset image size

This work was inspired by [67] where they used the computational efforts required using the full combination options of efficiency measures to seek the optimum design strategy for a prototypical single-family house. In this study, optimal design values for 18 EEMs were evaluated through DB. In comparison to the current building, the simulation results of the enhanced building based on the various parameters examined showed a noticeable difference. Having investigated the effect of each parameter as described previously, this section will analyse the overall effect of the combination of all measures listed in table 5.

The combination of variables yielded a significant total energy reduction of 67% with only 15 467 kW h being used annually. Generally, the simulation tends to follow same trend as the base case where the peak energy consumption falls in July–August months (hottest months in the KSA). Hence, largest reductions also happened during summer months (May–Oct), shown in figure 14. The cooling load of the existing building was responsible for 82% of the whole energy consumption. The significance of cooling is obvious since it was responsible to nearly 97% of total energy savings obtained in the improved design.

Zoom In Zoom Out Reset image size

The price of 1 kW h of electricity is 0.18 Saudi Riyal (equivalent to $0.0479) according to the current stated cost of power production in the KSA [68]. This energy reduction leads to 5746 SR being saved of the energy operation cost. Moreover, 19.35 ton of CO₂ emission would also be reduced when adopting the changes in the improved case as mentioned in table 5. Environmentally, the avoided GHG emissions can be expressed in the annual number of cars not used. To clarify, when the annual amount of CO₂ avoided is divided by the typical passenger vehicle emission which is about 4.6 metric tons of per year, 4.2 vehicles will not be used. This GHG analysis was conducted according to the United States Environmental Protection Agency [69].